CLOUDS. 441 



over, from llie fact that tbe stratum of cirrus is formed many liours, and 

 even days, before that of mimulus, especially in the equatorial regions, 

 and disappears after it. Without this distinction, we are obliged to call 

 the first stratum oi the 2yttlUuni cirrus, and the second cuimdus; but as, 

 under this state of strata, the form and i)hysical properties of cirrus and 

 cumulus change completely, confusion and errors continually result. 



Eegarding Howard's classification as a whole, while retaining the two 

 types of cirrus and cumulus, together with his two derivative clouds, 

 cirro-stratus and cirro-cumulus, I reject entirely" his stratus, niuibus, and 

 cumulo-stratus, as well as the strato-cumulus of Kiimtz, for the following 

 reasons : The stratus,* because it is not (according to Howard) a cloud 

 properly so called, but a mist, or lioar-frost, or the effect of optical illu- 

 sion, a cirrus, cirro-stratus, or cirro-cumiihis, seen in perspective near the 

 horizon; the nimhus, for the reason that the name is an inexact denom- 

 ination applied to an idea as vague as it is incorrect from the moment 

 that cumulus is not truly rainy, as far as it is found displayed, forming 

 a stratum as dense in appearance and below a second superior stratum 

 of cirrus, equally rainy; the cumulo-stratus, because it diifers in nothing 

 from cumulus, according to Howard's own definition, these two forms 

 Ijossessing in common the three fundamental characteristics of his cloud- 

 types and their derivatives, namely, horizontal bases, superior hemi- 

 spherical basins, and the ascending aggregation of their aqueous parti- 

 cles ; and, lastly, strato-cumulus, (Kiimtz's cloud of night,) because this 

 modifiCtation answers no better to clouds of night than Howard's stratus, 

 and because its other characteristics correspond to cumulo-stratus. 



On the other hand, as I have said before, I substitute for Qiimhis the 

 pallium, which I subdivide into ])allio-cirrus. and 2)allio-Gumulus, accord- 

 ing as its stratum is composed of cirrus or ctimulus. This term has the 

 triple advantage of embracing the character, the form, and the effect, as 

 the cirrus or cum^dus forms the rainy stratum. I introduce, in fine, the 

 definition of a second intermediate form, which can be rigorously distin- 

 guished from the preceding by the double relation of cause and effect. 

 This is the fracto-cumidus, fragments of cloud floating about without 

 determined form. Before their transformation into cumulus, they are 

 precipitated or detached from the inferior surface of. ^allio-cumidus, and 

 are sj)read out in horizontal bands at the top of the cumidus on the ap- 

 proach of gusts of wind. ThesQ fracto-cumuli differ from the cumulus 

 in that they have neither the horizontal base nor the sui)erior hemi- 

 spherical form, while they are not much extended ; but when they are 

 a little more increased we see at once forming at the center of the frag- 

 ment a space more dense and blackish than the rest, which gradually 

 settles until it constitutes the horizontal base of the cumidus, (vcl cu- 



* In Howard's Plate YI, j)ublislied iu Tilloch's Philosoiikical Magazine, in 1803, he rep- 

 resents tliis cloud as a mist spreading above a lake sui-rounded by hills. All succeeding 

 meteorologists have misunderstood this plate, and given the siraius as a baud spread 

 out at the horizon. 



